Intelligent motion stride estimation method and system

Abstract

The invention discloses a motion stride intelligent estimation method and system, and the method comprises the steps: calculating a first initial motion stride by a device end when a GPS is not started, and storing the first initial motion stride and a corresponding first stride state mark in a storage module, corresponding to a current motion type, of a server; the server side calculates a stride variance and a stride range according to the second movement stride array, and obtains an absolute value of a stride difference value between a stride mean value of the second movement stride array and the initial movement stride; and performing data fitting on the motion stride mean value and the first initial motion stride by using the fitting factor to obtain a second initial motion stride, and updating the initial motion stride by using the second initial motion stride and the corresponding second stride state flag. The accuracy of calculating the movement stride and the step number distance of the user is improved, and more personalized and accurate data information is provided for the movement of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent estimation of movement stride, and in particular to a method and system for intelligent estimation of movement stride. Background Art

[0002] There are two main types of stride length algorithms commonly used in wearable watches and wristbands. One is an estimation algorithm for scenarios where GPS is disabled, such as indoors or when the device itself lacks GPS and requires a mobile phone connection to use GPS data. The other is a calculation method for scenarios where GPS is enabled, such as outdoors. Stride length estimation for devices without GPS is significantly influenced by personal information such as height, and cannot be personalized based on the user's walking, running, posture, and habits. This results in lower accuracy and can significantly bias further distance traveled.

[0003] The accuracy of distance data with GPS function is relatively stable. Combined with the step count during exercise, it can meet the user's demand for accurate stride calculation. Currently, when the GPS function is activated (outdoors), the stride algorithm of watches and bracelets uses the distance measured by GPS and the step count to calculate, which is highly accurate and can also provide data information for evaluating the user's exercise stride. However, when the GPS function is not available (or indoors), the stride calculated based on personal information such as height is less accurate and can easily cause large deviations in distance calculations. Summary of the Invention

[0004] Existing smart wearable devices have low stride calculation accuracy in scenarios without GPS function, which causes large deviations in distance calculation.

[0005] To address the above problems, a method and system for intelligent estimation of exercise stride is proposed. The decision tree model is used on the device side to classify exercise and obtain the exercise type. The device side estimates the initial stride for each exercise type when GPS is not enabled (or used for the first time). The server side calculates and obtains the initial exercise stride when GPS is enabled. The initial exercise stride of the exercise type estimated by the device side is automatically corrected and updated. The corrected stride information will be associated with the corresponding exercise type and stored on the server side for use by the device side when GPS is not enabled (or in scenarios where GPS cannot be used) in the future. This improves the accuracy of the calculation of the user's exercise stride and step distance, and provides more personalized and accurate data information for user exercise.

[0006] In a first aspect, a method for intelligently estimating movement stride length adopts the following technical solution, including:

[0007] Step 100: The device obtains the current exercise type. When GPS is not enabled, the device obtains user personal information and calculates the current peak-to-valley amplitude average and cadence average of the motion sensor, obtains a sensor weighting factor and a cadence weighting factor, and calculates a first initial exercise stride using the personal information, the peak-to-valley amplitude average, the cadence average, the sensor weighting factor, and the cadence weighting factor. The first initial exercise stride and a corresponding first stride status flag are stored in a storage module corresponding to the current exercise type on the server.

[0008] Step 200: The server side obtains a first motion stride array corresponding to the motion type calculated, obtained, and transmitted by the device side when GPS is turned on, removes the maximum and minimum values in the first motion stride array, obtains a second motion stride array, calculates a stride variance and a stride range based on the second motion stride array, and if the stride range is less than or equal to a first specified threshold and the stride variance is less than or equal to a second specified threshold, obtains an absolute value of a stride difference between a stride mean of the second motion stride array and the first initial motion stride;

[0009] Step 300: If the absolute value of the stride difference is greater than or equal to a third specified threshold, obtain the mean motion stride and the first initial motion stride corresponding to the motion type, and use the fitting factor to perform data fitting on the mean motion stride and the first initial motion stride to obtain the second initial motion stride, and use the second initial motion stride and the corresponding second stride status flag to update the initial motion stride.

[0010] In conjunction with the method for intelligently estimating movement stride length described in the first aspect of the present invention, in a first possible implementation manner, the method for intelligently estimating movement stride length further includes:

[0011] Step 400: Obtain the exercise type of the user. If GPS is not enabled during the current exercise, obtain the stride status flag of the current exercise type.

[0012] Step 500: If the stride state flag is marked as the second stride state flag, read the corresponding second initial motion stride and use it as the initial motion stride of the motion type to estimate the stride.

[0013] In conjunction with the first possible implementation of the first aspect of the present invention, in a second possible implementation, step 100 includes:

[0014] Step 110: The device obtains the peak-to-valley amplitude, cadence, and number of steps of the user's exercise status;

[0015] Step 120: Determine the user's current exercise type based on the peak-to-valley amplitude value, cadence, and number of steps to obtain the user's current exercise type.

[0016] In conjunction with the second possible implementation manner of the first aspect of the present invention, in a third possible implementation manner, step 120 includes:

[0017] Step 121: If the peak-to-valley amplitude value is greater than or equal to the first peak-to-valley threshold, the cadence is greater than or equal to the first cadence threshold, and the number of steps is greater than or equal to the first number threshold, then determine that the current exercise type is a sprint exercise type;

[0018] Step 122: If the peak-to-valley amplitude value is greater than or equal to the second peak-to-valley threshold and less than the first peak-to-valley threshold, the cadence is greater than or equal to the second cadence threshold and less than the first cadence threshold, and the number of steps is greater than or equal to the second step number threshold, then the current exercise type is determined to be a regular running exercise type.

[0019] In combination with the second possible implementation manner of the first aspect of the present invention, in a fourth possible implementation manner, step 120 further includes:

[0020] Step 123: If the peak-to-valley amplitude value is greater than or equal to the third peak-to-valley threshold and less than the second peak-to-valley threshold, the cadence is greater than or equal to the third cadence threshold and less than the second cadence threshold, and the number of steps is greater than or equal to the third step number threshold, then determine that the current exercise type is a brisk walking exercise type;

[0021] Step 124: If the peak-to-valley amplitude value is greater than or equal to the fourth peak-to-valley threshold and less than the third peak-to-valley threshold, the cadence is greater than or equal to the fourth cadence threshold and less than the third cadence threshold, and the number of steps is greater than or equal to the fourth step number threshold, then it is determined that the current exercise type is a regular walking exercise type.

[0022] In conjunction with the second possible implementation manner of the first aspect of the present invention, in a fifth possible implementation manner, after step 120, step 100 further includes:

[0023] Step 130: Obtain the user's height information, compare the height information with a threshold, and determine the value of the sensor weighting factor based on the comparison result;

[0024] Step 140: Obtain the average cadence value, compare the average cadence value with a threshold, and determine the value of the cadence weighting factor according to the comparison result.

[0025] In conjunction with the first possible implementation of the first aspect of the present invention, in a sixth possible implementation, step 300 includes:

[0026] Step 310: Obtain the first initial movement stride and the first stride state flag of the corresponding movement type from the storage module;

[0027] Step 320: Replace the first initial motion stride with the second initial motion stride, update the first stride state flag to the second stride state flag, and store the flag.

[0028] In a second aspect, a system for intelligently estimating movement stride length adopts the intelligent movement stride length estimation method, comprising:

[0029] Device side;

[0030] Server side;

[0031] The device end is communicatively connected with the server end;

[0032] The device is configured to obtain a current exercise type, calculate a first exercise stride array corresponding to the exercise type when GPS is turned on, and obtain user personal information and calculate a current peak-to-valley amplitude average and a current cadence average of the motion sensor when GPS is not turned on, obtain a sensor weighting factor and a cadence weighting factor, calculate a first initial exercise stride using the personal information, the peak-to-valley amplitude average, the cadence average, the sensor weighting factor, and the cadence weighting factor, and store the first initial exercise stride and a corresponding first stride status flag in a storage module corresponding to the current exercise type on the server;

[0033] The server side is used to remove the maximum and minimum values in the first motion stride array transmitted from the device side, obtain a second motion stride array, calculate the stride variance and stride range based on the second motion stride array, if the stride range is less than or equal to the first specified threshold and the stride variance is less than or equal to the second specified threshold, then obtain the absolute value of the stride difference between the stride mean of the second motion stride array and the initial motion stride, if the absolute value of the stride difference is greater than or equal to the third specified threshold, then obtain the motion stride mean and the first initial motion stride corresponding to the motion type, and use the fitting factor to perform data fitting on the motion stride mean and the first initial motion stride to obtain the second initial motion stride, and use the second initial motion stride and the corresponding second stride status flag to update the initial motion stride.

[0034] In combination with the intelligent motion stride estimation system described in the second aspect of the present invention, in a first possible implementation manner, the server side is also used to: obtain the motion type of the user during exercise; if GPS is not turned on for the current exercise, obtain the stride status flag of the current motion type; if the stride status flag is marked as the second stride status flag, read the corresponding second initial motion stride and estimate the stride as the initial motion stride of the motion type.

[0035] In conjunction with the first possible implementation manner of the second aspect of the present invention, in a second possible implementation manner, the device side is further configured to:

[0036] Obtain the peak-to-valley amplitude value, cadence, and number of steps of the user's exercise state, determine the user's current exercise type based on the peak-to-valley amplitude value, cadence, and number of steps, and obtain the user's current exercise type.

[0037] The method and system for intelligently estimating exercise stride described in the present invention are implemented, and a decision tree model is used on the device side to perform exercise classification and obtain the exercise type. The device side estimates the initial stride for each exercise type when GPS is not enabled (or used for the first time). The server side calculates and obtains the initial exercise stride when GPS is enabled, and automatically corrects and updates the initial exercise stride of the exercise type estimated by the device side. The corrected stride information is associated with the corresponding exercise type and stored on the server side for use by the device side when GPS is not enabled (or in scenarios where GPS cannot be used) in the future. This improves the accuracy of calculating the user's exercise stride and step distance, and provides more personalized and accurate data information for user exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the solutions in this application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of this application. Obviously, the drawings described below are some embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0039] Figure 1 A flowchart of an embodiment of an intelligent estimation method for movement stride length according to the application;

[0040] Figure 2 yes Figure 1 A flowchart of a specific implementation of S100;

[0041] Figure 3 yes Figure 2 A flowchart of a specific implementation of S120;

[0042] Figure 4 yes Figure 3 A flowchart of a specific implementation method after S122;

[0043] Figure 5 yes Figure 2 A flowchart of a specific implementation method after S120;

[0044] Figure 6 yes Figure 1 A flowchart of a specific implementation of S300;

[0045] Figure 7 This is a flowchart of another embodiment of a method for intelligently estimating movement stride length according to the present application;

[0046] Figure 8 This is a schematic diagram of the module structure of an intelligent motion stride estimation system applied for;

[0047] Figure 9 This is a schematic diagram of personal information storage on the server side of an applied-for intelligent motion stride estimation system. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0049] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0051] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0052] Existing smart wearable devices have low stride calculation accuracy in scenarios without GPS function, which causes large deviations in distance calculation.

[0053] To solve the above problems, a method and system for intelligent estimation of motion stride length are proposed.

[0054] In the first aspect, a method for intelligently estimating movement stride length adopts the following technical solutions, such as Figure 1 , Figure 1 This is a flowchart of an embodiment of an intelligent motion stride estimation method of the application, including: S100, the device end 601 obtains the current motion type, if the GPS is not turned on, then obtains the user's personal information and calculates the current peak-to-valley amplitude average and the cadence average of the motion sensor, obtains the sensor weighting factor and the cadence weighting factor, and uses the personal information, the peak-to-valley amplitude average, the cadence average, the sensor weighting factor and the cadence weighting factor to calculate the first initial motion stride, and stores the first initial motion stride and the corresponding first stride status flag in the storage module corresponding to the current motion type of the server, such as Figure 9 , Figure 9 This is a schematic diagram of personal information storage on the server side 602 of the application for an intelligent motion stride estimation system.

[0055] In this embodiment, the motion sensor configured on the device end 601 may be a 3-axis acceleration sensor, or a 6-axis motion sensor or a 9-axis motion sensor.

[0056] In a preferred embodiment, Figure 2 , Figure 2 yes Figure 1 Flowchart of a specific implementation of S100; S100 includes: S110, the device end 601 obtains the peak-to-valley amplitude value peak_am, the cadence S and the number of steps count of the user's motion state; S120, judges the user's current motion type according to the peak-to-valley amplitude value peak_am, the cadence S and the number of steps count, and obtains the user's current motion type.

[0057] In a preferred embodiment, Figure 3 , Figure 3 yes Figure 2 A flowchart of a specific implementation of S120 in the figure; S120 includes: S121. If the peak-to-valley amplitude value peak_am is greater than or equal to the first peak-to-valley threshold d0, the cadence S is greater than or equal to the first cadence threshold S0, and the number of steps count is greater than or equal to the first number of steps threshold num0, then the current motion type is determined to be a sprint motion type (motion=0); S122. If the peak-to-valley amplitude value peak_am is greater than or equal to the second peak-to-valley threshold d1 and less than the first peak-to-valley threshold d0, the cadence S is greater than or equal to the second cadence threshold S1, and the number of steps count is greater than or equal to the second number of steps threshold num1, then the current motion type is determined to be a regular running motion type (motion=1).

[0058] In a preferred embodiment, Figure 4 , Figure 4 yes Figure 3 A flowchart of a specific implementation method after S122 in the figure; S120 also includes: S123, if the peak-to-valley amplitude value peak_am is greater than or equal to the third peak-to-valley threshold d2 and less than the second peak-to-valley threshold d1, the cadence S is greater than or equal to the third cadence threshold S2 and less than the second cadence threshold S1, and the number of steps count is greater than or equal to the third step threshold num2, then the current motion type is determined to be a brisk walking motion type (motion=2); S124, if the peak-to-valley amplitude value peak_am is greater than or equal to the fourth peak-to-valley threshold d3 and less than the third peak-to-valley threshold d2, the cadence S is greater than or equal to the fourth cadence threshold S3 and less than the third cadence threshold S2, and the number of steps count is greater than or equal to the fourth step threshold num3, then the current motion type is determined to be a conventional walking motion type (motion=3).

[0059] In this embodiment, the principles of the motion type classification model are as follows:

[0060] When the peak-to-valley amplitude value peak_am is greater than or equal to d0 (d0 is generally 3100), the step frequency is greater than or equal to the preset threshold value S0, and at the same time the number of steps is greater than or equal to num0 (num0 is generally 15-20), it is determined to be a sprint motion type, and motion=0 is stored in the server side 602. When the previous S does not hold, the next S determination is continued. When the peak-to-valley amplitude value peak_am is greater than or equal to d1 (d1 is generally 2800) and less than d0, the step frequency S is greater than or equal to the preset threshold value S2 and less than S1, and at the same time the number of steps is greater than or equal to num1 (num1 is generally 15-20), it is determined to be a regular running motion type, and motion=1 is stored in the server side 602. The subsequent S determination principle is the same, where d2 is generally 2500, d3 is 1800, and the values of num0, num1, num2, and num3 can be the same or different.

[0061] In a preferred embodiment, Figure 5 , Figure 5 yes Figure 2 Flowchart of a specific implementation method after S120 in the embodiment; after S120, S100 also includes: S130, obtaining the user's height information H, comparing the height information H with a threshold, and determining the value of the sensor weighting factor a1 according to the comparison result; S140, obtaining the cadence average S, comparing the cadence average S with the threshold, and determining the value of the cadence weighting factor a2 according to the comparison result.

[0062] In this embodiment, if the GPS function is not enabled on the device 601, the initial stride estimation S is performed. This module is mainly based on the personal information (height), peak-to-valley amplitude value peak_am average and average cadence sent to the device 601 by the server 602 (APP). The personal information includes but is not limited to the registered account, skin color code, height (H), weight (G), age (Age), gender (Gen), resting heart rate, etc. The height (H) is used here, and the unit time window of peak_am and average cadence is generally selected as 6s. The cadence is the average value of the number of steps within the unit time window. The calculation method of the first initial exercise stride Step_int is as shown in formula (1):

[0063]

[0064] in, It is the mean of the peak-to-valley amplitude value peak_am. When the height (H) is not greater than 155 cm, the sensor weighting factor a1 is 0.2036, otherwise a1 is 0.2607. When the average step frequency S is not greater than 120 times / minute, the step frequency weighting factor a2 is -0.1825, otherwise a2 is -0.2268. At this time, the first step state flag flag1 of the initial stride Step_int is 0.

[0065] When the GPS is turned on, the device 601 calculates and obtains the first step array. When the movement distance is greater than or equal to L0 cm (L0 is usually 300) and the number of steps is greater than or equal to N0 (N0 is usually 4-5), the first movement stride Step_move in the first step array is:

[0066]

[0067] Among them, Step_move is the first movement stride, unit is cm, Dist is the movement distance of the user in GPS startup mode, unit is cm, and Num_step is the statistical number of steps of the user in GPS startup mode.

[0068] In this embodiment, a plurality of first movement steps Step_move constitute a first step array.

[0069] S200, the server side 602 obtains the first motion stride array of the corresponding motion type calculated, obtained and transmitted by the device side 601 when the GPS is turned on, removes the maximum and minimum values in the first motion stride array, obtains the second motion stride array, calculates the stride variance and the stride range based on the second motion stride array, and if the stride range is less than or equal to the first specified threshold and the stride variance is less than or equal to the second specified threshold, then obtains the absolute value of the stride difference between the stride mean of the second motion stride array and the first initial motion stride.

[0070] S300. If the absolute value of the stride difference is greater than or equal to the third specified threshold, the mean motion stride and the first initial motion stride of the corresponding motion type are obtained, and the fitting factor is used to perform data fitting on the mean motion stride and the first initial motion stride to obtain the second initial motion stride, and the initial motion stride is updated using the second initial motion stride and the corresponding second stride status flag.

[0071] In a preferred embodiment, Figure 6 , Figure 6 yes Figure 1 A flowchart of a specific implementation of S300; S300 includes S310, obtaining the first initial motion stride and the first stride status flag of the corresponding motion type in the storage module; S320, replacing the first initial motion stride with the second initial motion stride, updating the first stride status flag to the second stride status flag, and storing them.

[0072] In this embodiment, when calculating the second initial exercise stride, personal information is stored in the server side 602, and the personal information includes but is not limited to the registration account, skin color code, height (H), weight (G), age (Age), gender (Gen), resting heart rate, exercise type, initial stride, initial stride status and exercise stride, etc.

[0073] For the first step array of length N1 (generally N1 is 10 to 15), the number of first motion strides is greater than or equal to a specified threshold (the specified threshold can be 8). The maximum and minimum values in the first step array are removed, and the stride range and stride variance in the first step array are calculated. The stride range is less than or equal to the first specified threshold delta1 (delta1 can be 3 to 5) and the stride variance is less than or equal to the second specified threshold delta2 (delta2 can be 1.3 to 1.7). If the conditions are met, the absolute value of the difference between the stride mean of the second motion stride array and the stride of the initial motion stride is obtained, and it is determined whether the absolute value of the difference between the stride mean of the remaining N1-2 motion steps and the stride of the first initial motion stride is greater than or equal to the third specified threshold delta3 (generally delta3 can be 12 to 16). If the conditions are met, the initial motion stride is updated according to formula (3):

[0074] Step_int=α*Step_int+(1-α)*mean(Step_move) (3),

[0075] Where α and β are data fitting factors, for example, α is set to 0.15. The updated initial stride will replace the original initial exercise stride and be stored in the personal information module of the server 602. Its status is marked as flag = 1 (the initial value of the flag is 0). When the device is subsequently used in an exercise type without the GPS function activated, if flag = 1, the stride will directly use the stored value of the system, thereby realizing the system's personalized intelligent stride calculation for the user.

[0076] In a preferred embodiment, Figure 7 , Figure 7 This is a flowchart of another embodiment of a method for intelligently estimating motion stride of the present application; the method for intelligently estimating motion stride also includes: S400, obtaining the type of motion of the user during motion, and if GPS is not turned on for the current motion, obtaining the stride status flag of the current motion type; S500, if the stride status flag is marked as a second stride status flag, reading the corresponding second initial motion stride, and estimating the stride as the initial motion stride of the motion type.

[0077] The device side 601 uses a decision tree model to classify exercise and obtain the exercise type. The device side 601 estimates the initial stride for each exercise type when GPS is not enabled (or used for the first time). The server side 602 calculates and obtains the initial exercise stride when GPS is enabled, and automatically corrects and updates the initial exercise stride of the exercise type estimated by the device side 601. The corrected stride information will be associated with the corresponding exercise type and stored in the server side 602 for use by the device side 601 when GPS is not turned on in the future (or in scenarios where GPS cannot be used). This improves the accuracy of the calculation of the user's exercise stride and step distance, and provides more personalized and accurate data information for user exercise.

[0078] Second, as Figure 8 , Figure 8 This is a schematic diagram of the module structure of an intelligent motion stride estimation system for the application; an intelligent motion stride estimation system, employing an intelligent motion stride estimation method, comprising a device end 601 and a server end 602; the device end 601 being communicatively connected to the server end 602; the device end 601 being configured to obtain a current motion type, calculate a first stride array when GPS is enabled, and obtain user personal information and calculate the current peak-to-valley amplitude average and stride frequency average of the motion sensor when GPS is disabled, obtain a sensor weighting factor and a stride frequency weighting factor, calculate a first initial motion stride using the personal information, the peak-to-valley amplitude average, the stride frequency average, the sensor weighting factor, and the stride frequency weighting factor, and store the first initial motion stride and the corresponding first stride status flag in a storage module corresponding to the current motion type on the server;

[0079] The server side 602 is used to remove the maximum and minimum values in the first motion stride array, obtain the second motion stride array, calculate the stride variance and stride range based on the second motion stride array, and if the stride range is less than or equal to the first specified threshold and the stride variance is less than or equal to the second specified threshold, then obtain the absolute value of the stride difference between the stride mean of the second motion stride array and the initial motion stride; if the absolute value of the stride difference is greater than or equal to the third specified threshold, then obtain the motion stride mean and the first initial motion stride of the corresponding motion type, and use the fitting factor to perform data fitting on the motion stride mean and the first initial motion stride to obtain the second initial motion stride, and use the second initial motion stride and the corresponding second stride status flag to update the initial motion stride.

[0080] The server side 602 in this application may include three parts: a storage module, a stride intelligent calculation module, a gateway module, and an APP module. The server is mainly responsible for the storage, management, and distribution of user information. The gateway module is responsible for screening request information. The APP module is responsible for obtaining user input information, launching some functions, uploading information to the server, and distributing information to the device side 601, playing a key role in data collection and transmission.

[0081] The device end 601 may include a motion sensor data module, a data preprocessing module, a motion type classification module, a step counting module, a GPS function judgment module, an initial stride calculation module, a distance parameter module, a distance and step counting judgment module, and a motion stride calculation module.

[0082] Motion sensor data includes, but is not limited to, 3-axis acceleration data, divided into three channels: X, Y, and Z, with a frequency of 25 Hz. S104 is the data preprocessing module, which performs filtering, baseline removal, and peak and valley detection, and extracts step counts based on the periodicity of the 3-axis acceleration data.

[0083] Specifically, the acceleration X, Y, and Z axis signals are subjected to SG filtering respectively to extract the signal baseline, wherein the SG filter order includes but is not limited to 1st order, and the filter buffer length includes but is not limited to 1 times the sensor sampling rate. The baseline signals of the X, Y, and Z axis signals are removed respectively, and the AC signals are extracted by low-pass and high-pass filtering in turn, wherein the low-pass and high-pass filters include but are not limited to 2nd order, and the types include but are not limited to IIR filtering. The peak-valley threshold method is used to extract the characteristic points of the peaks and valleys of the X, Y, and Z three-channel AC signals, and the extraction method includes but is not limited to the peak-valley threshold method. The signal with the largest peak-to-valley amplitude value in the X, Y, and Z three-axis signals is selected as the target signal for motion classification and statistical step counting, wherein the peak-to-valley amplitude value is recorded as peak_am.

[0084] Furthermore, the server side 602 is also used to: obtain the exercise type of the user during exercise; if the GPS is not turned on during the current exercise, obtain the stride status flag of the current exercise type; if the stride status flag is marked as the second stride status flag, read the corresponding second initial exercise stride and use it as the initial exercise stride of the exercise type to estimate the stride.

[0085] Furthermore, the device end 601 is further used to obtain the peak-to-valley amplitude value, cadence and number of steps of the user's motion state, judge the user's current motion type based on the peak-to-valley amplitude value, cadence and number of steps, and obtain the user's current motion type.

[0086] The method and system for intelligently estimating motion stride of the present invention utilizes a decision tree model to classify motion and obtain motion types through the device end 601. When GPS is not enabled (or used for the first time), the device end 601 estimates the initial stride for each motion type. The server end 602 calculates and obtains the initial motion stride when GPS is enabled, and automatically corrects and updates the initial motion stride of the motion type estimated by the device end 601. The corrected stride information is associated with the corresponding motion type and stored in the server end 602 for use by the device end 601 when GPS is not enabled (or in scenarios where GPS cannot be used) in the future. This improves the accuracy of calculating the user's motion stride and step distance, and provides more personalized and accurate data information for user motion.

[0087] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for intelligently estimating a movement stride, comprising: Step 100: The device obtains the current exercise type. When GPS is not enabled, the device obtains user personal information and calculates the current peak-to-valley amplitude average and cadence average of the motion sensor. A sensor weighting factor and a cadence weighting factor are obtained. A first initial exercise stride is calculated using the personal information, the peak-to-valley amplitude average, the cadence average, the sensor weighting factor, and the cadence weighting factor. The first initial exercise stride and a corresponding first stride status flag are stored in a storage module corresponding to the current exercise type on the server. Step 200: The server side obtains a first motion stride array corresponding to the motion type calculated, obtained, and transmitted by the device side when GPS is turned on, removes the maximum and minimum values in the first motion stride array, obtains a second motion stride array, calculates a stride range and a stride variance based on the second motion stride array, and if the stride range is less than or equal to a first specified threshold and the stride variance is less than or equal to a second specified threshold, obtains the absolute value of the stride difference between the stride mean of the second motion stride array and the first initial motion stride; Step 300: If the absolute value of the stride difference is greater than or equal to a third specified threshold, obtain the mean motion stride and the first initial motion stride corresponding to the motion type, and use the fitting factor to perform data fitting on the mean motion stride and the first initial motion stride to obtain the second initial motion stride, and use the second initial motion stride and the corresponding second stride status flag to update the initial motion stride.

2. The method for intelligent estimation of movement stride according to claim 1, characterized in that: The method for intelligently estimating movement stride also includes: Step 400: Obtain the exercise type of the user. If GPS is not enabled during the current exercise, obtain the stride status flag of the current exercise type. Step 500: If the stride state flag is marked as the second stride state flag, read the corresponding second initial motion stride and use it as the initial motion stride of the motion type to estimate the stride.

3. The method for intelligently estimating movement stride according to claim 2, wherein: The step 100 includes: Step 110: The device obtains the peak-to-valley amplitude, cadence, and number of steps of the user's exercise status; Step 120: Determine the user's current exercise type based on the peak-to-valley amplitude value, cadence, and number of steps to obtain the user's current exercise type.

4. The method for intelligently estimating movement stride according to claim 3, wherein: The step 120 includes: Step 121: If the peak-to-valley amplitude value is greater than or equal to the first peak-to-valley threshold, the cadence is greater than or equal to the first cadence threshold, and the number of steps is greater than or equal to the first number threshold, then determine that the current exercise type is a sprint exercise type; Step 122: If the peak-to-valley amplitude value is greater than or equal to the second peak-to-valley threshold and less than the first peak-to-valley threshold, the cadence is greater than or equal to the second cadence threshold and less than the first cadence threshold, and the number of steps is greater than or equal to the second step number threshold, then the current exercise type is determined to be a regular running exercise type.

5. The method for intelligent estimation of movement stride according to claim 4, characterized in that: The step 120 further includes: Step 123: If the peak-to-valley amplitude value is greater than or equal to the third peak-to-valley threshold and less than the second peak-to-valley threshold, the cadence is greater than or equal to the third cadence threshold and less than the second cadence threshold, and the number of steps is greater than or equal to the third step number threshold, then determine that the current exercise type is a brisk walking exercise type; Step 124: If the peak-to-valley amplitude value is greater than or equal to the fourth peak-to-valley threshold and less than the third peak-to-valley threshold, the cadence is greater than or equal to the fourth cadence threshold and less than the third cadence threshold, and the number of steps is greater than or equal to the fourth step number threshold, then it is determined that the current exercise type is a regular walking exercise type.

6. The method for intelligently estimating movement stride according to claim 2, wherein: After step 120, step 100 further includes: Step 130: Obtain the user's height information, compare the height information with a threshold, and determine the value of the sensor weighting factor based on the comparison result; Step 140: Obtain the average cadence value, compare the average cadence value with a threshold, and determine the value of the cadence weighting factor according to the comparison result.

7. The method for intelligently estimating movement stride according to claim 2, wherein: The step 300 includes: Step 310: Obtain the first initial movement stride and the first stride state flag of the corresponding movement type from the storage module; Step 320: Replace the first initial motion stride with the second initial motion stride, update the first stride state flag to the second stride state flag, and store the flag.

8. An intelligent sports stride estimation system, using the intelligent sports stride estimation method according to any one of claims 1 to 7, characterized in that: include: Device side; Server side; The device end is communicatively connected with the server end; The device is configured to obtain a current exercise type, calculate a first exercise stride array corresponding to the exercise type when GPS is turned on, and obtain user personal information and calculate a current peak-to-valley amplitude average and a current cadence average of the motion sensor when GPS is not turned on, obtain a sensor weighting factor and a cadence weighting factor, calculate a first initial exercise stride using the personal information, the peak-to-valley amplitude average, the cadence average, the sensor weighting factor, and the cadence weighting factor, and store the first initial exercise stride and a corresponding first stride status flag in a storage module corresponding to the current exercise type on the server; The server side is used to remove the maximum and minimum values in the first motion stride array transmitted from the device side, obtain a second motion stride array, calculate the stride variance and stride range based on the second motion stride array, if the stride range is less than or equal to the first specified threshold and the stride variance is less than or equal to the second specified threshold, then obtain the absolute value of the stride difference between the stride mean of the second motion stride array and the initial motion stride, if the absolute value of the stride difference is greater than or equal to the third specified threshold, then obtain the motion stride mean and the first initial motion stride corresponding to the motion type, and use the fitting factor to perform data fitting on the motion stride mean and the first initial motion stride to obtain the second initial motion stride, and use the second initial motion stride and the corresponding second stride status flag to update the initial motion stride.

9. The intelligent motion stride estimation system according to claim 8, characterized in that: The server side is also used to: obtain the exercise type of the user during exercise; if GPS is not turned on for the current exercise, obtain the stride status flag of the current exercise type; if the stride status flag is marked as the second stride status flag, read the corresponding second initial exercise stride and use it as the initial exercise stride of the exercise type to estimate the stride.

10. The intelligent motion stride estimation system according to claim 9, characterized in that: The device is further configured to: Obtain the peak-to-valley amplitude value, cadence, and number of steps of the user's exercise state, determine the user's current exercise type based on the peak-to-valley amplitude value, cadence, and number of steps, and obtain the user's current exercise type.

Images